clinical study preoperative strength training for elderly...

Clinical StudyPreoperative Strength Training for Elderly PatientsAwaiting Total Knee Arthroplasty

D. M. van Leeuwen,1,2 C. J. de Ruiter,1 P. A. Nolte,3 and A. de Haan1,2

1 MOVE Research Institute Amsterdam, Faculty of Human Movement Sciences, VU University Amsterdam,Van der Boechorststraat 9, 1081 BT, Amsterdam, The Netherlands

2 Institute for Biomedical Research into HumanMovement and Health, Manchester Metropolitan University, Manchester M1 5GD, UK3Department of Orthopedics, Spaarne Hospital, Spaarnepoort 1, 2134 TM Hoofddorp, The Netherlands

Correspondence should be addressed to A. de Haan; [email protected]

Received 4 September 2013; Revised 12 December 2013; Accepted 31 December 2013; Published 13 February 2014

Academic Editor: Jiu-jenq Lin

Copyright © 2014 D. M. van Leeuwen et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Objective. To investigate the feasibility and effects of additional preoperative high intensity strength training for patients awaitingtotal knee arthroplasty (TKA). Design. Clinical controlled trial. Patients. Twenty-two patients awaiting TKA. Methods. Patientswere allocated to a standard training group or a group receiving standard training with additional progressive strength training for6weeks. Isometric knee extensor strength, voluntary activation, chair stand, 6-minute walk test (6MWT), and stair climbing wereassessed before and after 6weeks of training and 6 and 12weeks after TKA. Results. For 3 of the 11 patients in the intensive strengthgroup, training load had to be adjusted because of pain. For both groups combined, improvements in chair stand and 6MWTwereobserved before surgery, but intensive strength training was not more effective than standard training. Voluntary activation did notchange before and after surgery, and postoperative recovery was not different between groups (𝑃 > 0.05). Knee extensor strength ofthe affected leg before surgery was significantly associated with 6-minute walk (𝑟 = 0.50) and the stair climb (𝑟− = 0.58, 𝑃 < 0.05).Conclusion. Intensive strength training was feasible for themajority of patients, but there were no indications that it is more effectivethan standard training to increase preoperative physical performance. This trial was registered with NTR2278.

1. Introduction

Knee osteoarthritis (OA) is a degenerative joint diseasewhich is characterized by a gradual loss of cartilage [1] andcan result in pain, limited physical performance, and lowerquality of life [2]. If conservative treatment is ineffective,patients may decide to undergo a total knee arthroplasty(TKA), which can significantly reduce knee pain and canincrease physical performance in patients with severe OA[1]. For patients undergoing TKA, the isometric strength ofthe knee extensors was shown to decrease by up to 60%four weeks after surgery, and this decrease was accompaniedby decreases in the ability to voluntarily activate the kneeextensor muscles [3]. Even after six months to thirteen yearsfollowing TKA, the strength of the knee extensor musclesat involved side remains 12–30% lower than the uninvolvedside, and strength almost never matched values for healthy

controls [4]. This postoperative weakness has importantconsequences for activities of daily life, because knee extensorstrength is strongly related to functional performance, suchas walking and stair climbing [5] especially after TKA [6].There are indications that preoperative strength is relatedto postoperative abilities [7, 8]. Intensive strength trainingafter TKA has shown to be beneficial for decreasing painand improving strength and physical performance whencompared to usual care [2].Multiple studies have investigatedthe effect of preoperative strength training on postoperativerecovery [9–15]. However, few of these studies reportedsignificant increases in preoperative strength following thetraining. Reviewing these studies, it is clear that the intensityof training, when documented, was either rather low [10,13–15] or not progressively increased [13] or the numberof sessions was too small to produce significant trainingeffects [9]. Progressive, high intensity strength training is

Hindawi Publishing CorporationRehabilitation Research and PracticeVolume 2014, Article ID 462750, 9 pageshttp://dx.doi.org/10.1155/2014/462750

2 Rehabilitation Research and Practice

recommended to increase muscle strength [16]. Because thepreoperative training period is typically rather short (thetime between the decision for TKA and the actual surgeryis typically 4 to 8 weeks), a high intensity and progressiveloading may be needed to increase preoperative strength andperformance and therefore promote postoperative recovery.However, it is unclear if this type of training is feasible in thispatient group, since pain may be a limiting factor.

The aims of the present study were to investigate thefeasibility and the effects of additional preoperative highintensity strength training for elderly patients awaiting TKAcompared to standard preoperative training in a pilot study.Wehypothesised that preoperative intensive strength trainingwould lead to increases in strength and performance beforesurgery. We hypothesised that increases in strength wereprimarily caused by improved voluntary activation, becausethe first adaptations to strength training are primarily neural[17] and training time is limited.

2. Methods

2.1. Participants. All patients above 55 years awaiting TKAin the Spaarne Hospital in Hoofddorp were consideredcandidates for the present study andwere asked to participate.Patients were excluded if they had (1) American Society ofAnesthesiologists (ASA) score >2 [18], (2) contraindicationsfor training the lower limbs, or (3) contraindications forelectrical stimulation (unstable epilepsy, cancer, skin abnor-malities, or having a pacemaker).

All patients had at least 1 year symptoms of severeosteoarthritis of the knee (Kellgren and Lawrence [19] grades3 and 4). For additional exercise, patients were asked abouttheir physical activity at each measurement occasion. Thepatients did not perform strength training before inclusionin this study. No severe coexisting diseases were present.Therefore, we do not expect a limiting effect on function orexercise responses of our participants.

2.2. Sample Size. Isometric knee extension strength of thesurgical leg before TKA was defined as the primary outcomevariable for the power analysis. The effect size for strengthtrainingwith patients having osteoarthritis has been reportedto be 0.35 [20] and 0.30 for preoperative training [14].Because the control group also received therapy, we used aneffect size of 0.20. For 0.8 power, 𝛼 = 0.05 and assuming acorrelation of 0.85 between repeated measurements, a totalof 18 participants was needed to assess significant differencesbetween groups over time. Because 4 participants droppedout before the second measurement, four additional patientswere included and in total 22 patients were enrolled in thestudy.

2.3. Randomization and Blinding. Participants were random-ized in a 1 : 1 ratio (parallel design) to either the standardtreatment or standard treatment with additional strengthtraining. A research nurse approached potential candidatesby phone, generated the random allocation sequencewith useof custom software, enrolled patients, and assigned them to

the interventions. Randomization was done by minimizationof gender and age (median age of patients on the waiting list).After the inclusion of 15 patients, 2 participants had droppedout and two patients received the intervention instead ofstandard training and the ratio between strength trainingand standard training was 10/3. To increase comparabilitybetween groups, the remaining 7 patients were allocated tothe standard training group. The principal investigator (DL)was blinded duringmeasurements, but not during analyses ofthe results. The participants and therapists were not blinded.

2.4. Surgical Procedure. Patients underwent an uncementedTKA (mobile bearing total knee prosthesis, LCS Complete,Depuy, Warsaw, Indiana, United States) with standardizedperioperative protocol and the same surgical technique. Thesurgical technique consisted of a midline incision with aflexed knee,medial arthrotomy, and bone cuts withMilestoneinstruments without the use of tourniquet or drains. Periop-erative antibiotics (Kefzol 1 gram i.v.) and antithrombotics(Fraxiparine 0.3mL i.m.) were used.The patients were mobi-lized the first day postoperatively. On average the patients leftthe hospital the 4th postoperative day.The surgeries were per-formed by experienced orthopedic surgeons (>50 TKA peryear) and patients received protocolized inpatient physicaltherapy. The VU Medical Centre Medical Ethics Committeeand the local ethics committee of the Spaarne Hospitalapproved the study, and all participants signed informedconsent and the rights of the subjects were protected.

2.5. Intervention. Patients were allocated to standard treat-ment or received standard treatment with additional strengthtraining (Figure 1). The standard training group receivedtreatment according to guidelines from theDutch associationof orthopaedics [21] and the Dutch physiotherapy association(KNGF) [22] for training patients with OA.Therapy includedinformation and advice, exercise of activities of daily life,training of walking with aids, maintenance of mobility, andaerobic training (walking, cycling), but the patients in thisgroup were not allowed to perform resistance training. Theintensive strength training group received the same treatmentas the standard training group, with additional intensivestrength training, consisting of a progressive strength pro-gram targeting the lower limbs. Table 1 shows exercises, sets,and repetitions. We abstained from 1 RM testing to minimizepain sensations, because pain could lead to premature endingof the training. Instead, the training weights were adjustedto the abilities of the patients in relation to the numberof repetitions. For the first training (3 × 15 repetition),patients were asked to perform the maximum number ofrepetitions with the selected weight. If either more or lessthan 15 repetitions were performed, the weight for the nextset was adjusted with ∼3% per repetition. For example, if apatient could perform 22 repetitions with 30 kg, the weightwas increased with 7 (22 − 15 repetitions) ∗ 3% to 36.6 kg.Dumbbells or plates were used for small increments. Toensure progressive overload, repetitions decreased during theprogram, and the weights were increased when the numberof repetitions decreased (∼3% per repetition). For the squat

Rehabilitation Research and Practice 3

Randomized:N = 22

Lost to follow-up (N = 4)

Completed testing (analyzed)

T1:N =

T2:N =

T3:N =

T4:N =

Completed testing (analyzed)

T1:N =

T2:N =

T3:N =

T4:N =

Lost to follow-up (N = 2):

Standard training group:N = 11 Strength training group:N = 11

-

-T2: gout (N = 1)

-T4: 2nd TKA (N = 1)

-T2: pain during training (N = 1)

-T2: surgery cancelled becauseof less pain (N = 1)

-T3: hematoma (N = 1)

11 (8)

8 (8)

7 (7)

7 (7)

11 (10)

10 (10)

10 (10)

9 (9)

- T2: ASA2 ⇢ ASA3 (N = 1)

Figure 1: Flowchart of inclusion and follow-up in the two training groups.

Table 1: Exercises, sets and repetitions for the strength traininggroup.

Week 1 Week 2 Week 3 Week 4 Week 5 Week 6Leg press1-leg 3 × 15 3 × 12 4 × 12 3 × 10 4 × 10 4 × 8

Step up 1-leg 3 × 15 3 × 12 4 × 12 3 × 10 4 × 10 4 × 8Squat 3 × 15 3 × 12 4 × 12 3 × 10 4 × 10 4 × 8

Legextension1-leg

3 × 15 3 × 12 4 × 12 3 × 10 4 × 10 4 × 8

exercise, intensity was increased by the increasing the rangeof motion before using dumbbells. Both the uninvolved andthe involved limb were trained, and the weight was adjustedto abilities. The patients trained two to three times per week.In addition, a home program consisting of step-up and squatexercises was performed two to three times per week by thestrength training group. In case of pain or other discomfort,the program was modified, but the intensity stayed as high aspossible. After surgery, no interventions were applied; bothgroups received standard care including strength training.13 physiotherapy centres participated by complying with thetraining program. 22 patients entered the study. Figure 1shows allocation and follow-up.

2.6. Measures. All measurements were performed at theSpaarne Hospital before training (T1), after 6 weeks oftraining (T2, the week before TKA), 6 weeks after surgery(T3), and 12 weeks after surgery (T4).

2.6.1. Feasibility. The feasibility was evaluated by checkingtraining logs for adherence. Physiotherapists were instructedto note alterations of the training program. If trainingintensity was progressively increased and all exercises wereexecuted, the program was considered feasible. The numberand contents of the training sessions for the control groupwere also monitored by checking training logs.

2.6.2. TorqueMeasurements . Measurement of the contractileproperties of the knee extensor and flexor muscles took placeon a custom-made adjustable dynamometer. The lower legwas tightly strapped to a force transducer (KAP-E, 2 kN,A.S.T., Dresden, Germany),mounted to the frame of the chairabout 25 cm distally of the knee joint. Participants sat in thedynamometer with a hip angle of 80∘ (0∘ is full extension),firmly attached to the seat with straps at the pelvis to preventextension of the hip during contraction and a strap at thechest. All measurements were performed on both legs at aknee angle of 60∘ (0∘ is full extension), during isometriccontraction. The nonsurgical leg was measured first to getaccustomed to the procedures and electrical stimulation (see


below). Force data were digitized (1 kHz), filtered with a 4thorder bidirectional 150Hz Butterworth low-pass filter, andstored on a PC for offline analysis. Force signals were cor-rected for gravity: the average force applied by the weight ofthe limbwas set at zero. Torque was calculated bymultiplyingforce with the distance between the force transducer andthe knee joint. After 3 submaximal attempts, participantswere asked to perform at least 3 maximal isometric kneeextensions and flexions, and more if torque increased morethan 10%, with at least two minutes of rest in betweenattempts. Maximal Voluntary Torque was defined as thehighest torque recorded.

2.6.3. Electrical Stimulation. Constant current electricalstimulation (pulse width 200𝜇s) was applied through self-adhesive surface electrodes (Schwa-Medico, Leusden, TheNetherlands) by a computer-controlled stimulator (modelDS7A, Digitimer Ltd., Welwyn Garden City, UK). The distalelectrode (8 × 13 cm) was placed over the medial partof the quadriceps muscle just above the patella and theproximal electrode (8 × 13 cm) over the lateral portionof the muscle to prevent inadvertent stimulation of theadductors. Before placing the electrodes the skin in thearea of the electrodes was shaved. The stimulation currentwas increased until force in response to doublet stimulation(two pulses at 100Hz) levelled off. After assessing maximaldoublet force, the stimulation intensity was lowered and set toproduce 50% of the maximal doublet force. This stimulationintensity ensured that a substantial amount of muscle masswas stimulated but significantly reduced discomfort at thesame time [23]. Voluntary activation was calculated withuse of the superimposed twitch technique. In short, upon amaximal voluntary contraction, a superimposed doublet wasdelivered to the muscle. Two seconds after each contraction,a (potentiated) doublet was delivered to the relaxed muscleto calculate voluntary activation with use of the followingequation:

Voluntary activation (%)

= 1 − (superimposed force

potentiated resting doublet) ∗ 100%

(1)

(see [23, 24]).

2.6.4. Functional Tasks. A 5-time sit-to stand test was per-formed with the arms folded in front of the chest. Patientswere instructed to stand up and sit down as quickly aspossible. The six-minute walk test (6MWT) was used toquantify walking ability. Participants walked back and forthover 30 meters as many times as possible for a period of6 minutes at their own pace, in a 60-meter-long corridor.The score recorded was the total distance travelled during 6minutes. Participants were instructed to “walk as quickly andsafely as you can for 6 minutes.”

To investigate stair-climbing, the time required to ascend9 steps, turn around, and descend 9 steps was used. Partic-ipants were allowed to use the handrail and instructed to“walk as quickly and safely”. All tests except the 6MWT were

repeated twice, and the fastest time was used for analysis.The6MWTand the stair climb test arewidely used as specific teststo quantify functional performance in patients [6, 25–28].

2.6.5. Quality of Life and Physical Activity. Quality of life wasassessed with the Western Ontario and McMaster Univer-sities Osteoarthritis Index (WOMAC). The WOMAC ques-tionnaire is used to obtain pain, stiffness, and functioningspecifically for patients with OA. Scores were transformed toa 0 to 100 scale, where a 100 score signifies the best quality oflife.

2.7. Statistics. Data are presented as mean ± SD. An ANOVArepeated measure was used to assess differences betweenthe patient groups over time with a Bonferroni post-hoccorrection. Two separate analyses were performed. The firstanalysis was done with preoperative data of patients withdata on T1 and T2 (𝑁 = 18, T1 and T2) because theprimary aim was to study effects of training on preoperativestrength and performance. A second analysis was done onall complete data sets (T1–T4; 𝑁 = 16) to investigatepostoperative recovery (T3 and T4). Because not all patientswere randomized, a per-protocol analysis was performed. Achi-square test was used to investigate differences in genderat baseline.Other baseline characteristicswere analysed usingthe Kruskal-Wallis Test.

Effect size was calculated by subtracting the mean pre-post (T1-T2) change in the standard group from themeanpre-post change in the intensive training group, divided by thepooled pre-test standard deviation [29].

Pearson’s correlation coefficient was used to investigaterelationships between normally distributed variables. Thelevel of significance for all tests was set at 0.05 and all analyseswere performed with SPSS (version 16.0, SPSS Inc.).

3. Results

3.1. Feasibility. Twenty-two patients were recruited betweenOctober 2010 and December 2011. Figure 1 shows a flowchartof allocation and follow-up. All participants in the strengthtraining group completed preoperative training, and therewas one dropout in the standard training group. Fourparticipants did not complete the 2nd preoperative test dueto various reasons (Figure 1). Only data were analysed frompatients who completed testing at T2.

Eight out of 11 patients in the strength training groupcompleted training without adaptations. For 3 patients, smalladjustments were made in intensity due to pain, to preventpremature ending of the training. Patients in the strengthtraining group completed 12 ± 2 training sessions (range 11–17), and patients in the standard training group completed11 ± 4 sessions (range 4–16).

In a pilot study, split squats were included in the trainingprogram, but too many patients reported pain during thisexercise. Also reduction in range of motion in knee exten-sions and leg press showed to be an effective way to reducepain, while maintaining a high training intensity.


Table 2: Characteristics of patients of the two training groups anddrop-outs.

Strengthtraining(𝑁 = 10)

Standardtraining(𝑁 = 8)

Drop-outs(𝑁 = 4) 𝑃

Sex(men/women)a 7/3 4/4 1/3 0.30

Age(years)b 71.8 (7.5) 69.5 (7.1) 73.3 (3.4) 0.33

BMI(kg/m2)b 27.9 (4.6) 27.9 (3.1) 26.3 (2.1) 0.71

aDifferences tested using 𝜒2 test.bPresented as mean (standard deviation), differences tested using Kruskall-Wallis Test.

Table 2 shows baseline characteristics for the patients inthe strength training group, the standard training group whocompleted testing at T2, and the patients that dropped out.There were no significant differences between the groups.

3.2. Pre-Surgery Effects

3.2.1. Strength Measures. Table 3 shows average values forstrength measures. Before surgery there were no main effectsof group or time: at baseline and T2, there were no significantdifferences in strength measures between groups and nochanges in time for the total group.The effect size of maximalvoluntary knee extension strength was 0.11. The post-hocpower was 0.87. There were also no significant interactionsbetween group and time for any strength measure duringthis six-week preoperative training period. Strength trainingdid not lead to increases in maximal knee extension torque(Table 3), voluntary activation, or doublet torque comparedto the standard training group. At T1 and T2, the affected legwas not weaker than the unaffected leg and also voluntaryactivation was not different between both legs. The patientswho dropped out before T2 did not have a significantlylower knee extension strength of the affected leg than thepatients who completed testing at T2 (98Nm versus 113Nm,𝑃 = 0.61). The percentage of men and women in the twogroups differed. Therefore, we also compared knee extensiontorque between the two groups at baseline with correction forgender. Both without (𝑃 = 0.929) and with correction forgender (𝑃 = 0.769), there was no difference in maximal kneeextension torque at baseline.

3.2.2. Functional Tasks. Before surgery (from T1 to T2) therewere nomain effects of “group,” but there weremain effects of“time” for chair stand and 6MWT. For both groups combinedchair stand (−1.1 s, 𝑃 = 0.003) and 6MWT (25m, 𝑃 =0.013) significantly improved before surgery (Table 3) andthere was a trend for improvement in voluntary knee flexionstrength of the affected side (3.4Nm, 𝑃 = 0.090). Therewere no significant interactions between “time” and “group,”indicating that any changes over time were similar betweengroups.

3.3. Post-Surgery Effects

3.3.1. Strength Measures. After surgery there were no maineffects of “group”. There was a main effect of “time” formaximal knee extension torque, doublet torque, andmaximalknee flexion torque of the affected knee. Maximal torque ofthe knee extensors and doublet torque significantly decreasedfrom T2 to T3 (6 weeks after surgery) and subsequentlysignificantly increased from T3 to T4 (12 weeks after surgery,𝑃 < 0.05, Table 3). Knee flexor torque significantly increasedfrom T3 to T4. At T4, maximal torque for knee extension anddoublet torquewere still between 20 and 30% lower comparedwith their preoperative values at T2, whereas maximal torquefor knee flexion was back to baseline levels.

An unexpected finding was that there was a significantinteraction between maximal torque of the knee flexors andgroup. Post-hoc testing indicated that maximal torque of theknee flexors decreased in the standard training comparedto the intensive training groups between T2 and T3. Asexpected, doublet torque and knee extensor torque werelower for the affected side compared to the unaffected side onT3 and T4 and knee flexor torque was lower at T3 only (𝑃 <0.05) compared to the unaffected side. Voluntary activationdid not change after surgery.

3.3.2. Functional Tasks. After surgery, there were no maineffects of “group”, but there were main effects of “time” forseveral variables. Six weeks after surgery (T3), stair climbingtime increased compared to T2 for both groups combined.From T3 to T4, significant main effects of time were presentfor chair stand, stair climb, 6MWT, andWOMAC score (𝑃 <0.05, Table 3), without any significant interaction betweengroup and time, again indicating that any changes over timewere similar between groups.

3.4. Relationships between Quadriceps Strength and PhysicalPerformance. Table 4 shows Pearson’s correlation coefficientsbetween maximal knee extension strength and chair, stairclimb, and 6MWT performance at the four moments oftesting. Only after surgery, maximal knee extension strengthwas related to chair stand (𝑟2 = 0.27 and 𝑟 = 0.31, 𝑃 < 0.05).Stair climb performance was related to maximal torque ofboth legs on all occasions (𝑟2 between 0.28 and 0.55,𝑃 < 0.05)and 6MWT was significantly related to strength on T2, T3,and T4 (𝑟2 > 0.25, 𝑃 < 0.05). In general, relationshipsbetween voluntary knee extensor strength and the functionaltests became stronger over time.

4. Discussion

The main findings of the present study were that intensivestrength training is feasible for the majority of the patientsawaiting TKA, but that there are no indications that thisintensive strength training is more effective than a standardtraining.The feasibility and pre- and postoperative effects willbe separately discussed.


Table 3: Strength measures, functional tasks, and WOMAC scores before (T1, T2) and after (T3, T4) surgery.

T1 (𝑁 = 10/8) T2 (𝑁 = 10/8) T3 (𝑁 = 10/7) T4 (𝑁 = 9/7)

MVT extension (Nm)Affected side STR 106 ± 45 111 ± 50

∗63 ± 30

∗76 ± 34

STAND 121 ± 52 121 ± 50 70 ± 35 97 ± 40

Unaffected side STR 116 ± 47 123 ± 47 116 ± 44 118 ± 43

STAND 137 ± 59 139 ± 57 128 ± 65 138 ± 56

Doublet Torque (Nm)Affected side STR 49 ± 13 50 ± 16

∗34 ± 10

∗39 ± 12

STAND 51 ± 19 48 ± 17 35 ± 13 39 ± 14


STAND 50 ± 15 50 ± 16 50 ± 17 50 ± 13

VA (%)Affected side STR 79 ± 13 78 ± 15 79 ± 9 80 ± 10

STAND 80 ± 13 85 ± 8 84 ± 4 90 ± 8


STAND 84 ± 12 85 ± 10 88 ± 6 91 ± 6

MVT flexion (Nm)Affected side STR 40 ± 22 43 ± 19

†37 ± 18

∗42 ± 17

STAND 46 ± 25 50 ± 24 36 ± 16 50 ± 23


STAND 57 ± 33 55 ± 30 55 ± 30 55 ± 26

Chair stand test (s) STR 12.6 ± 2.6∗11.3 ± 2.1 13.3 ± 3.4

∗11.8 ± 1.8

STAND 12.3 ± 2.7 11.4 ± 1.8 12.5 ± 2.5 10.8 ± 1.5

Stair climb test (s) STR 12.4 ± 3.1 11.6 ± 3.4∗20.9 ± 10.8

∗12.8 ± 3.4

STAND 12.9 ± 3.8 12.4 ± 3.3 17.6 ± 7.5 14.1 ± 0

6MWT (m) STR 453 ± 81∗471 ± 92 380 ± 109

∗456 ± 62

STAND 460 ± 52 493 ± 55 440 ± 87 513 ± 97

WOMAC score (points) STR 64 ± 11 65 ± 20 70 ± 16∗83 ± 15

STAND 67 ± 11 67 ± 8 79 ± 11 93 ± 4

MVT:Maximal voluntary torque; VA: voluntary activation; 6MWT: six-minute walk test;WOMAC:McMaster Universities Osteoarthritis Index; STR: strengthtraining group; STAND: standard training group.The numbers of patients in the intervention and standard training groups are displayed at the different times.Values represent mean ± standard deviation. ∗Significantly different compared to previous measurement for both groups combined (𝑃 < 0.05). †Significantdifference for groups between T3 and T2 (𝑃 = 0.043).

Table 4: Pearson correlation coefficients between maximal kneeextension strength and functional tests.

Chairstand test

Stairclimb test

6-minutewalk test

Affectedside

T1 −0.03 −0.53∗ 0.41T2 −0.32 −0.58∗ 0.50∗

T3 −0.56∗ −0.68∗ 0.76∗

T4 −0.56∗ −0.74∗ 0.86∗

Unaffectedside

T1 −0.17 −0.59∗ 0.46T2 −0.32 −0.64∗ 0.54∗

T3 −0.47 −0.59∗ 0.66∗

T4 −0.52∗ −0.73∗ 0.77∗∗

𝑃 < 0.05.

4.1. Feasibility. One of the aims of the present study wasto investigate the feasibility of additional preoperative highintensity strength training for elderly patients awaiting TKA.In this training group, no patients dropped out because of theintervention. For 3 out of 11 patients, changes in the programhad to be made because of pain or discomfort, but for theother 8 patients the training program could be performed

without alterations. Although the groups were of limited size,intensive strength training seems feasible, at least for patientswith ASA 1 or 2.

4.2. Pre-Surgery Effects. The effect size of the training onstrength was small, 0.11, and not significant. This was notin line with our expectations, but it might be explained bythe relatively short training time. Six weeks of training twotimes per week might not be enough to significantly increasestrength in patients with end-stageOA, even if a high trainingintensity is used. In a systematic review investigating effectsof strength training in OA patients, positive effects have beenreported on strength, performance, and pain compared tocontrol groups [30].The average duration of the studies in thisreview was 9 months. Longer interventions may be neededto significantly increase preoperative strength and physicalperformance.

Therewere no differences in strength between the affectedand the unaffected leg before surgery, although a difference instrength is often observed [14, 31, 32].Thismight be explainedby the fact that 2 patients were having a second TKA at a laterstage and 4 patients already had an earlier TKA.This indicatesthat the nonsurgical leg was not “unaffected” in all patients.


The finding that strength training did not increase pre-operative strength or promote postoperative outcome is inline with the majority of earlier studies [9, 10, 12–14]. Inthe present study, there were improvements in chair standand the 6MWT for the entire group before surgery. It isimportant to note that both groups in the current studyreceived training. In the absence of training, strength andperformance often decline in the preoperative period [9,14, 15], which was not the case in the current study. Thestandard training group in the present study underwentaerobic training (walking and cycling), balance training, andtraining of activities of daily life, such as chair rises andbasic step training. In many other studies no exercise isprescribed during the preoperative period for a control group[9–15]. Because both groups trained, this may not only haveprevented the decline as is seen in many other studies duringthe preoperative phase, but it also seems to suggest that theexact content of the training program is less relevant duringa short preoperative phase. This finding is in line with theresults of a recent study in which a control group improvedwalking and stair climbing after 6 weeks of nonspecificupper-body strength training [33]. There are no indica-tions in the present study that additional heavy resistancetraining is superior to a program of more general aerobictraining including some functional (strength demanding)tasks.

4.3. Post-Surgery Effects. The recovery of voluntary torque,stair climb, and walking ability at T4 was comparable to twoearlier studies [34, 35], but somewhat lower than reported byothers [14, 32].There was a significant interaction (𝑃 = 0.043)between group and time for maximal torque of the knee flex-ors from T2 to T3. This interaction was probably not causedby the intensive strength training, because no interaction waspresent before surgery, and the preoperative training programwas primarily focused on the knee extensors. Therefore,we consider this to be a sporadic finding. There were noother significant interactions between group and time aftersurgery.

4.4. Voluntary Activation. Before surgery, there were nodifferences in voluntary activation between the surgical andnonsurgical leg. As stated before, the lack of changes mightbe caused by an earlier or a future TKA of the nonsurgicalleg. There were also no changes in voluntary activationafter training and after surgery. The absence of changes involuntary activation is not in line with two earlier studies[3, 36] that measured lower activation 4 weeks after surgery,but in accordance with two other studies in which no changeswere found 12 weeks after surgery [37, 38]. The differentfindings regarding changes in voluntary activation may beexplained by differences in timing of the measurements aftersurgery among studies.Thirty threemonths after surgery, sig-nificant increases in voluntary activation have been observedcompared to before surgery [31]. Voluntary activation maydecrease the first weeks after surgery and improve on a longerterm.

4.5. Relationships between Quadriceps Strength and PhysicalPerformance. The relationships between strength and phys-ical performance and the observation that relationships arestronger later after surgery are in line with other studies[6, 32]. This may indicate that knee extension strength is animportant factor for performance, especially in later stages ofrecovery. Consequently, postoperative strength training mayimprove functional recovery, which is in line with earlierresearch [2].

4.6. Clinical Relevance and Limitations. A major strength ofthe current study compared to other studies is that preop-erative training had a relative high intensity and loads wereprogressively increased. For patients, the results of the presentstudy indicate that it is unnecessary to subject patients tointensive training before TKA. A limitation is the low samplesize in this study. Especially when studying postoperativeeffects, a larger sample size would be needed. It is, however,unlikely that preoperative strength trainingwould be effectiveto promote recovery after surgery compared to standardpreoperative training, because neither significant effects nortrends for superior effects of strength training were observedbefore surgery. Another limitation of the present study is alack of randomization and the lack of blinding for therapistsand patients.

5. Conclusion

We conclude that intensive strength training is feasible forthe majority of the patients awaiting TKA. There were noindications that this intensive strength training is more effec-tive than a standard training with respect to maximal kneeextensor strength, voluntary activation, and performance infunctional tests.

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper.

Acknowledgment

The authors would like to thank Jeanette Verhart for herefforts including the participants.

References

[1] S. M. Seed, K. C. Dunican, and A. M. Lynch, “Osteoarthritis: areview of treatment options,” Geriatrics, vol. 64, no. 10, pp. 20–29, 2009.

[2] S. C. Petterson, R. L. Mizner, J. E. Stevens et al., “Improvedfunction from progressive strengthening interventions aftertotal knee arthroplasty: a randomized clinical trial with animbedded prospective cohort,” Arthritis Care and Research, vol.61, no. 2, pp. 174–183, 2009.

[3] J. E. Stevens, R. L. Mizner, and L. Snyder-Mackler, “Quadricepsstrength and volitional activation before and after total kneearthroplasty for osteoarthritis,” Journal of Orthopaedic Research,vol. 21, no. 5, pp. 775–779, 2003.


[4] W. Meier, R. Mizner, R. Marcus, L. Dibble, C. Peters, andP. C. Lastayo, “Total knee arthroplasty: muscle impair-ments, functional limitations, and recommended rehabilitationapproaches,” Journal of Orthopaedic and Sports Physical Ther-apy, vol. 38, no. 5, pp. 246–256, 2008.

[5] M. R. Maly, P. A. Costigan, and S. J. Olney, “Determinants ofself-report outcome measures in people with knee osteoarthri-tis,” Archives of Physical Medicine and Rehabilitation, vol. 87, no.1, pp. 96–104, 2006.

[6] Y. Yoshida, R. L. Mizner, D. K. Ramsey, and L. Snyder-Mackler,“Examining outcomes from total knee arthroplasty and therelationship between quadriceps strength and knee functionover time,” Clinical Biomechanics, vol. 23, no. 3, pp. 320–328,2008.

[7] J. R. Jaggers, C. D. Simpson, K. L. Frost et al., “Prehabilitationbefore knee arthroplasty increases postsurgical function: a casestudy,” Journal of Strength andConditioning Research, vol. 21, no.2, pp. 632–634, 2007.

[8] R. L. Mizner, S. C. Petterson, J. E. Stevens, M. J. Axe, andL. Snyder-Mackler, “Preoperative quadriceps strength predictsfunctional ability one year after total knee arthroplasty,” Journalof Rheumatology, vol. 32, no. 8, pp. 1533–1539, 2005.

[9] D. S. Rooks, J. Huang, B. E. Bierbaum et al., “Effect of preop-erative exercise on measures of functional status in men andwomen undergoing total hip and knee arthroplasty,” ArthritisCare and Research, vol. 55, no. 5, pp. 700–708, 2006.

[10] J. A. Rodgers, K. L. Garvin, C. W.Walker, D. Morford, J. Urban,and J. Bedard, “Preoperative physical therapy in primary totalknee arthroplasty,” Journal of Arthroplasty, vol. 13, no. 4, pp. 414–421, 1998.

[11] J. Crowe and J. Henderson, “Pre-arthroplasty rehabilitationis effective in reducing hospital stay,” Canadian Journal ofOccupational Therapy, vol. 70, no. 2, pp. 88–96, 2003.

[12] D. D. D’Lima, C. W. Colwell Jr., B. A. Morris, M. E. Hardwick,and F. Kozin, “The effect of preoperative exercise on totalknee replacement outcomes,” Clinical Orthopaedics and RelatedResearch, no. 326, pp. 174–182, 1996.

[13] L. A. Beaupre, D. Lier, D. M. Davies, and D. B. C. Johnston,“The effect of a preoperative exercise and education programon functional recovery, health related quality of life, and healthservice utilization following primary total knee arthroplasty,”Journal of Rheumatology, vol. 31, no. 6, pp. 1166–1173, 2004.

[14] R. Topp, A. M. Swank, P. M. Quesada, J. Nyland, and A.Malkani, “The effect of prehabilitation exercise on strength andfunctioning after total knee arthroplasty,” PM and R, vol. 1, no.8, pp. 729–735, 2009.

[15] A. M. Swank, J. B. K. Kachelman, B. Wendy et al., “Prehabil-itation before total knee arthroplasty increases strength andfunction in older adults with severe osteoarthritis,” Journal ofStrength and Conditioning Research, vol. 25, no. 2, pp. 318–325,2011.

[16] American College of Sports Medicine position stand, “Progres-sion models in resistance training for healthy adults,”Medicine& Science in Sports & Exercise, vol. 41, no. 3, pp. 687–708, 2009.

[17] D. A. Gabriel, G. Kamen, and G. Frost, “Neural adaptationsto resistive exercise: mechanisms and recommendations fortraining practices,” Sports Medicine, vol. 36, no. 2, pp. 133–149,2006.

[18] A. S. O. Anesthesiologists, “New classification of physicalstatus,” Anesthesiology, vol. 24, p. 111, 1963.

[19] J. H. Kellgren and J. S. Lawrence, “Radiological assessment ofosteo-arthrosis,” Annals of the rheumatic diseases, vol. 16, no. 4,pp. 494–502, 1957.

[20] M. Fransen and S.McConnell, “Exercise for osteoarthritis of theknee,” Cochrane Database of Systematic Reviews, no. 4, 2008.

[21] NOV, Richtlijn diagnostiek en behandeling van heup-enknieartrose, 2007.

[22] E. M. H. M. Vogels, “KNGF-richtlijn Atrose heup-knie,” Neder-lands Tijdschrift Voor Fysiotherapie, vol. 115, no. 1, 2005.

[23] D. M. van Leeuwen, C. J. de Ruiter, and A. de Haan, “Effectof stimulation intensity on assessment of voluntary activation,”Muscle Nerve, vol. 45, no. 6, pp. 841–848, 2012.

[24] J. P. Folland and A. G. Williams, “Methodological issues withthe interpolated twitch technique,” Journal of Electromyographyand Kinesiology, vol. 17, no. 3, pp. 317–327, 2007.

[25] R. L. Mizner and L. Snyder-Mackler, “Altered loading duringwalking and sit-to-stand is affected by quadriceps weaknessafter total knee arthroplasty,” Journal of Orthopaedic Research,vol. 23, no. 5, pp. 1083–1090, 2005.

[26] A. de Haan, C. J. de Ruiter, L. H. van Der Woude et al.,“Contractile properties and fatigue of quadriceps muscles inmultiple sclerosis,” Muscle Nerve, vol. 23, no. 10, pp. 1534–1541,2000.

[27] C. J. De Ruiter, B. G. M. Van Engelen, R. A. Wevers, and A.De Haan, “Muscle function during fatigue in myoadenylatedeaminase-deficient Dutch subjects,” Clinical Science, vol. 98,no. 5, pp. 579–585, 2000.

[28] K. H. Gerrits, M. J. Beltman, P. A. Koppe et al., “Isometricmuscle function of knee extensors and the relation with func-tional performance in patients with stroke,” Archives of PhysicalMedicine and Rehabilitation, vol. 90, no. 3, pp. 480–487, 2009.

[29] S. B. Morris, “Estimating effect sizes from pretest-posttest-control group designs,” Organizational Research Methods, vol.11, no. 2, pp. 364–386, 2008.

[30] A. K. Lange, B. Vanwanseele, and M. A. Fiatarone Singh,“Strength training for treatment of osteoarthritis of the knee:a systematic review,” Arthritis Care and Research, vol. 59, no. 10,pp. 1488–1494, 2008.

[31] A. Berth,D.Urbach, andF.Awiszus, “Improvement of voluntaryquadriceps muscle activation after total knee arthroplasty,”Archives of Physical Medicine and Rehabilitation, vol. 83, no. 10,pp. 1432–1436, 2002.

[32] R. L. Mizner, S. C. Petterson, and L. Snyder-Mackler, “Quadri-ceps strength and the time course of functional recovery aftertotal knee arthroplasty,” Journal of Orthopaedic and SportsPhysical Therapy, vol. 35, no. 7, pp. 424–436, 2005.

[33] C. McKay, H. Prapavessis, and T. Doherty, “The effect of aprehabilitation exercise program on quadriceps strength forpatients undergoing total knee arthroplasty: a randomizedcontrolled pilot study,” PM& R, vol. 4, no. 9, pp. 647–656, 2012.

[34] M. J. Bade, W. M. Kohrt, and J. E. Stevens-Lapsley, “Outcomesbefore and after total knee arthroplasty compared to healthyadults,” Journal of Orthopaedic and Sports Physical Therapy, vol.40, no. 9, pp. 559–567, 2010.

[35] D. Ouellet and H. Moffet, “Locomotor deficits before and twomonths after knee arthroplasty,” Arthritis Care and Research,vol. 47, no. 5, pp. 484–493, 2002.

[36] R. L. Mizner, S. C. Petterson, J. E. Stevens, K. Vandenborne,and L. Snyder-Mackler, “Early quadriceps strength loss aftertotal knee arthroplasty: the contributions ofmuscle atrophy andfailure of voluntarymuscle activation,” Journal of Bone and JointSurgery A, vol. 87, no. 5, pp. 1047–1053, 2005.


[37] D. Vahtrik, H. Gapeyeva, H. Aibast et al., “Quadriceps femorismuscle function prior and after total knee arthroplasty inwomen with knee osteoarthritis,” Knee Surgery, Sports Trauma-tology, Arthroscopy, vol. 20, no. 10, pp. 2017–2025, 2011.

[38] A. Berth, D. Urbach, W. Neumann, and F. Awiszus, “Strengthand voluntary activation of quadriceps femoris muscle in totalknee arthroplasty with midvastus and subvastus approaches,”Journal of Arthroplasty, vol. 22, no. 1, pp. 83–88, 2007.

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

clinical study preoperative strength training for elderly...

Documents